Collaborative Thinking

E-mail stress - The Tip Of The Attention Iceberg?

The broader issue the article below raises is whether this type of worker stress is limited to e-mail? As XML feeds, instant messaging, blogs, social networking, presence-enabled applications and other social tools invade the workplace - how fragmented and disjointed will workers feel as they triage a growing number of communication channels and collaboration spaces all demanding their attention.

I would not jump to any melodramatic conclusion - some people adapt well and actually thrive within event-driven environments. Other workers will struggle until they learn new ways to cope and leverage tools that help filter, sort and prioritize work activities and manage interruptions. Technology to fix the problems that technology causes always lags a bit. The trend, to some extent, is unavoidable - creating a market opportunity for software vendors, consultants and perhaps even therapists I imagine...

More than a third said they thought they checked their inbox every 15 minutes and 64 per cent said they looked more than once an hour. When researchers fitted monitors to their computers, workers were found to be viewing e-mails up to 40 times an hour. About 33 per cent said they felt stressed by the volume of e-mails and the need to reply quickly. A further 28 per cent said they felt “driven” when they checked messages because of the pressure to respond. Just 38 per cent of workers were relaxed enough to wait a day or longer before replying.

E-mail stress keeps workers on edge of inbox - Times Online

... data from AOL and Opinion Research Corporation shows that Americans are increasingly "addicted" to checking their e-mail, with 59 percent of portable device users checking every single time an e-mail comes in and 83 percent of survey respondents checking e-mail every day on vacation.

All of this e-mail checking is not only stressing us out, it's cutting down on our productivity. Humans aren't quite as adept at multitasking as we'd like to think we are, and it takes us a long time to refocus; Microsoft recently said that workers took an average of 15 minutes to get back to what they were working on after being interrupted by a phone call, e-mail, or IM. Dr. Renaud agrees. "The problem is that when you go back to what you were doing, you've lost your chain of thought and, of course, you are less productive," she said. "People's brains get tired from breaking off from something every few minutes to check emails. The more distracted you are by distractions, including email, then you are going to be more tired and less productive."

E-mail stress slowing down workers, say researchers

Analytics: The Unsung Hero Of Social Systems

Stream processing systems (and associated analytical components) will become a critical underpinning for much of what is talked about in terms of workstreaming, lifestreaming, attention streams, collective intelligence and so on. Discovering patterns across people, interactions, information, activities and social networks and assessing those relationships is difficult enough. It becomes even more challenging when you also want the results to be communicated in a manner that is contextual, relevant and sensitive to attention (and confidentiality) needs.

Of note are aggregation and correlation hubs that can also add value on top of the analytics. A good example is Feedburner which allows augmentation of information with additional insertions that can alter attention, influence participation and encourage action on the information being communicated. The result is a type of network effect. Augmentation could include insertion of tagging options, references to relevant communities or workspaces, real-time counts on how other people are treating that information, pointers to workspaces, as well as widgets. I've alluded to these trends in a recent presentation on collective intelligence (via SlideShare here) and the relevant slide below.

A near-perfect world would:

Take what's going on in my life...

Directory of Lifestreaming

I probably should lump all these into the Directory of Attention, but I’m not going to.

Don’t look for a definition of lifestreams on Wikipedia, because it will take you to a Final Fantasy VII page. The term actually goes back to at least 1997, when Eric Freeman and David Gelernter saw it “as a network-centric replacement for the desktop metaphor. As their project page (last updated in 2000) at Yale put it:

A lifestream is a time-ordered stream of documents that functions as a diary of your electronic life; every document you create and every document other people send you is stored in your lifestream.

Source: loose wire blog

Add more context about what I'm doing...

+

My Data Stream

After a year and a half of using social applications heavily, I recently had to revisit the plan to aggregate all my activity into one data stream.  As the calendar rolled to 2007, I kept wishing I could look at all my social activity from 2006 in context: time, date, type of activity, location, memory, information interest, and so on.  What was I bookmarking, blogging about, listening to, going to, and thinking about?  I still had the urge to have an information and online activity mash-up that would allow me to discover my own patterns and to share my activity across the web in one chronological stream of data (to start with anyway).

Source: Emily Chang - Blog: My Data Stream

In the context of my interaction patterns

+

/Message: Traffic And Flow

Social applications are -- at their basis -- a means for us to communicate. Not just point-to-point communication, as in email or in IM, but increasingly a more general communication from me to the network of others that believe that I matter. This is what blogging affords us, and Flickr streams, and even Twitter.

We are sending all sorts of traffic -- different sorts of messages -- flowing through the various implicit and explicit social networks that we define ourselves through, and through which we discover meaning, belonging, and insight.

This traffic flow -- made more liquid by RSS and instant messaging style real-time messaging -- is the primary dynamic that I believe we will see in all future social apps. Yes, we will want to have our traffic cached -- for search and analysis purposes -- but we will increasingly move toward a flow model: where the various bits that we craft and throw into the ether -- blog posts, calendar entries, photos, presence updates, whatever -- will be picked up by other apps, either to display them to us, or to make sense of them. We want to consolidate all into one flow -- a single time-stamped thread -- that all apps can dip into.

Source: /Message: Traffic And Flow

Correlate everything in an intelligent manner...

+

Complex Event Processing for EDA

Organizations are increasingly adopting EDA as a platform to effectively manage the increasing number of events generated from IT systems, business processes and physical sensors such as RFID. With this in mind, Oracle is developing new Complex Event Processing (CEP) capabilities that support high-volume, programmatic analysis of events to identify patterns and correlations across multiple heterogeneous event sources. These capabilities will complement Oracle's existing EDA offerings, such as Oracle Business Activity Monitoring, that provides real-time operational dashboards for tracking business key performance indicators, multi-channel alerting, and invoking automated or manual response actions.

Source: Oracle Unveils Next-Generation Architecture for Oracle® Fusion Middleware

Continue to analyze continuously - past and present...

+

Streaming Analytics vs. Perpetual Analytics (Advantages of Windowless Thinking)

The terms "streaming" and "perpetual" probably sound like the same thing to most people. However, in the context of intelligent systems, I think there is a big difference.

[Note: when I use the term "observation" below, feel free to think about this as a synonym for "transaction" or "record."]

Streaming analytics involves applying transaction-level logic to real-time observations. The rules applied to these observations take into account previous observations as long as they occurred in the prescribed window – these windows have some arbitrary size (e.g., last five seconds, last 10,000 observations, etc.).

Perpetual Analytics, on the other hand, evaluates every incoming observation against ALL prior observations. There is no window size. Recognizing how the new observation relates to all prior observations enables the publishing of real-time insight (i.e., The Data Finds the Data and the Relevance Finds the User).  And another unique property is Sequence Neutrality (i.e., future observations can affect earlier outcomes).

Source: Jeff Jonas: Streaming Analytics vs. Perpetual Analytics (Advantages of Windowless Thinking)

And discover what is important to me (even if I may not know it)...

+

Creators of "The Content Router for the Web"

The Real Time Matrix provides technology to find what is relevant to "Me" from the continuous flow of live, Web content and send it to me on my personal channel. And, the content I want finds me in real time on my cell phone, IM, browser, reader, PDA, or any other Internet-connected device.

It is available today through the free iJ.am ™ web site as well as a mashup service that you may incorporate into your own products.

Source: The Real Time Matrix#

Augment that information before you communicate it to me...

+

FeedFlare

Give your subscribers easy ways to email, tag, share, and act on the content you publish by including as many or few of the services listed below. FeedFlare places a simple footer at the bottom of each content item, helping you to distribute, inform and create a community around your content.

Introduction

The FeedFlare API (FlareAPI) allows anyone to extend our existing FeedFlare service. Provide new actions and incorporate outside services to make your content more interesting and engaging — both in your FeedBurner feed and on your website.

Source: FeedBurner - FeedFlare Developer Guide

+

Signal such information or messages relevant to my work context and focus...

The New Presence Model

New Presence is a user-centric view of presence.  Instead of merely reflecting the crude, device specific “availability awareness” of today, New Presence systems understand our context, relationships, wants and desires.  The New Presence model reflects the integrated conversation web we live in today.

The New Presence model has three building blocks: relationships, context, profile.  Each of these is a core component in a model which is fundamentally richer, and more user-driven than any presence model previously.

Source: “New Presence” and the Voice 2.0 Manifesto -- Alec Saunders .LOG

+

and, of course, in a manner that is aware of my attention priorities...

Source: My Attention Management System Conceptual Architecture

Augmenting XML Feeds

While this post talks about Feedbuner's FeedFlare capability and how use of the feature can improve interactivity, the concept of augmenting feeds with various "attention analytics" and "action options" applies to enterprise use of feeds. For instance, I might only scan a particular feed item but if I noticed that it is being aggressively bookmarked I might re-visit the feed item to understand "what all the fuss is about". So certain persistent analytics on what other people are doing with a feed item can grab my attention even if I had only given the article a cursory glance initially. Within an enterprise, this could be fine-tuned to include counts by role (e.g., architects) so that feeds sent to architects would have specific analytics pertaining to that community group. While most of the footers added to a feed are rather obvious (e-mail, tag, digg) - for an enterprise the options can be more application and process-oriented (perhaps initiating a workflow). As a feed passes through a feed management server, that server has the opportunity to profile and correlate the feed to other activities and intelligently supplement the feed with action steps that make feed consumption a bridge to other work activities. For instance, you might added an action item to post the feed item to a a library or discussion forum within a virtual workspace site. The idea of having a back-end that perpetually analyzes all feeds suggests that there is the opportunity for an intranet version of a Feedburner-like service with similar Feed Flare functionality. Developers could build a catalog of attention analytics and action options to avoid feed items being a terminal end-point - instead, users have a options that continue the flow, transitioning the next activity to the proper tool, application and so on.

I did an interview on Monday where the podcaster asked me how to make feeds "stickier". What he was actually asking was how to get readers more engaged with feed content: how can feeds be made more interactive? A lot of the thinking behind FeedFlare was that we needed a way to give publishers tools to increase the likelihood that readers would in fact engage. Clicking through to read a copy of the post they just read is unlikely to drive a lot of click activity. But clicking through to read the comments will. Bookmarking the post at del.icio.us will drive further activity, as will voting for the post at Digg. (And in those latter examples, they'll both increase secondary traffic growth, by building awareness of your content at those sites.) In other words, adding opportunities for the readers to do things other than just read a copy of the post goes a long way to increasing the probability that the readers will actually do something.

Too few publishers take advantage of the next logical step: building their own FeedFlare units to direct attention to other parts of the publisher's site. If you publish archives by category, why not give readers the ability to browse more articles like the one they just read by going to the category archive? Promoting an event? Do what the folks at TechPresident are doing and include a link to the event with every post:

That link gets seen by everyone subscribed to the feed, dramatically increasing the visibility of the Personal Democracy Forum event (disclosure: I'm speaking at PdF, and FeedBurner's a sponsor). Creating this FeedFlare takes less than five minutes, and it's then something you can share with anyone else who wants to support the event. (I won't go into all the variations here, but creating FeedFlares for fundraising, micro-sites for a specific function, etc., all make a ton of sense. You get the idea.) At this point, the feed is not just a way of distributing content, but is equally about driving awareness and delivering actions - just not all focused exclusively on the individual post.

Source: Burning Questions • Rick's Ruminations: Full Feeds

Poynter Online - EyeTrack07: The Myth of Short Attention Spans

The video is well-worth the time to watch if you are involved in user experience and content delivery or interface design. If you are interested in orchestrating information in a way that maintains reader attention, this information will also be quite valuable. The data from this story could be of interest to vendors in the XML syndication business as well - the tidbits of information could be helpful in how to display streams of feed items  more effectively. Some of the study data points are also relevant to design of blogs and wikis. 

EyeTrack07: The Myth of Short Attention Spans

Poynter unveiled the initial findings of its most recent study of reader behavior at the ASNE convention in Washington, D.C., this morning. Watch, listen and read about them here first.

You can't get much more basic than the lead finding of Poynter's EyeTrack07 study, presented this morning to the American Society of Newspaper Editors in Washington, D.C.

Readers select stories of particular interest and then read them thoroughly.
And there's a twist: The reading-deep phenomenon is even stronger online than in print.

At a time when readers are assumed to have short attention spans, especially those who read online, this qualifies as news.

Source: Poynter Online - EyeTrack07: The Myth of Short Attention Spans

Exploiting Attention Data Streams

HP continues to do interesting research around social software. The del.icio.us and add-a-tag effort is very intriguing - one of the challenges when designing socially-oriented systems that help people collectively share information (and rely on that information as a connection mechanism) is to include automation services that help the environment learn and adapt by observing and interpreting user behaviors. This particular effort focuses on tags with the assumption that such meta data can be used as indicators of a user's attention and when analyzed over time, interesting correlations can be made that can be used to add value to profiles and also suggest strong/weak ties to other individuals (i.e., attention data streams as basis for some type of social network relationship).

del.icio.us user profiles

I've been showing Elke Michlmayr's Add-A-Tag algorithm to folks at ICWSM. As she writes:

If a user's tagging data is treated as a continuous stream of information about a user's interests, it can be used to create a rich user profile. The profile should represent the most important parts of a user's behaviour. Both persistent long-term interests and transient short-term interests should co-exist in the profile.

Elke was an intern at HP where she did this work.

Source: Semantic Blogging Demonstrator

The Add-A-Tag algorithm: Learning adaptive user profiles from tagging data

Due to the high popularity of folksonomies, a large amount of metadata is available. This collaboratively created data is a valuable resource. If a user’s tagging data is treated as a continuous stream of information about a user's interests, it can be used to create a rich user profile. The profile should represent the most important parts of a users‘ behaviour. Both persistent long-term interests and transient short-term interests should co-exist in the profile.

Source: WIT - The Add-A-Tag algorithm

A cure for e-mail attention disorder? | CNET News.com

Applying models and practices from gaming environments as a means to improve how organizations communicate has some intriguing possibilities. This article examines the use of currency and some tenets from the field of economics (scarce resources) as a means to encourage better behavior and attention management. 

A cure for e-mail attention disorder?

Silicon Valley start-up develops system to help manage e-mail overload that borrows heavily from the virtual economies and currencies.
Images: Handling e-mail overload

By Daniel Terdiman
Staff Writer, CNET News.com

Published: February 28, 2007, 4:00 AM PST

TalkBack E-mail Print del.icio.us Digg this

Corporate managers concerned about the amount of time employees spend sifting though mountains of unwanted e-mail may soon have World of Warcraft to thank for providing a solution.

That's because a Silicon Valley start-up called Seriosity has come up with a system to help manage e-mail overload that borrows heavily from the virtual economies and currencies found in WoW and other large-scale online games.

Known as Attent, Seriosity's system is essentially a new currency--called the Serio--that corporate e-mail users spend to indicate a message's importance: the more important they think the message is, the more Serios they spend on it. Recipients keep the Serios in the messages they get.

Similarly, when someone receives a message with Serios attached, they can indicate how important they think it is by responding with an appropriate number: none or very few if they think the message wasn't valuable, an equal number if they want the sender to know they appreciated the message or more than the original number to show they agree that it truly was crucial.

But Serios is a currency, and therefore a scarce resource, so users get a limited amount. The idea is that they have to spend the currency wisely, always making sure they have enough to send more with future messages.

And while the system, strictly speaking, is enterprise software, it was directly inspired by the virtual economies of online games like WoW. There, players accumulate gold or platinum pieces or some other form of currency and can spend them on weapons, armor, dwellings and the like that themselves have real monetary value as demonstrated by what people will pay for them on auction and third-party Web sites.

Ultimately, the point of Serios is to help large enterprises manage their employees' attention.

Source: A cure for e-mail attention disorder? | CNET News.com

Presence: Thinking Way Beyond The Buddy List

Just some short-hand notes (not complete thinking) to capture thoughts-in-progress:

Presence & identity: At the core, presence is "you" but how many "you"'s are there? I have many personas: Principal Analyst at Burton Group, a technology geek to my circle of friends, father/husband to my family, a customer to some companies I deal with, etc. In some cases, I'm a made-up person (screen names on AOL or a member of some social networking site). It's important to not go too far down the road in terms of solving presence without due diligence regarding how it relates to identity and how identity is managed and secured.

Presence & location: Presence also touches upon where I am - my home office, on the road at a Starbucks, at a hotel, at a conference, etc. Location information is also important to consider when devising a strategy around presence.

Presence & environment: Somewhat related to location, presence could include insight to the environment around you - the capabilities and/or constraints of your computing environment (connectivity not good enough for video) as well as the form factor(s) you have at the moment (mobile device vs. a PC).

Presence & activity: Presence is influenced/impacted by the activity you are involved in as well as the activities of others. Certain activities preclude me from being available. Other activities might make me freely available. And there are all sorts of combinations in-between.

Presence & role: People wear many hats. There are default roles that might show up in a presence profile but perhaps not all possible permutations. There might be roles that are viewable through certain filters. For instance, someone might be considered a first responder in case of an emergency on their floor in a corporate office. Loading the filter for a "first responder role" into a presence system could display that view of presence information. Roles might also be tacit, my activities might grant me a role for some duration, or my expertise (know-how) or network strength (know-who) might make me "present" on-the-fly.

Presence & meta data: Presence can be assigned to artifacts. I might tag an artifact with meta-data that is resolved in real-time (e.g., author for a document, subject-matter expert) that might only display in a list of search results. Presence is attachable to any application element. Presence as "live" meta data (presence-enabled systems) is a good discussion to have with developers (workflow, etc).

Presence & availability: Availability is a subjective thing - always, sometimes, never, to everyone, to someone, to no one. If I say no one do I not mean my wife or boss. Can I delegate availability? How does availability relate intelligently to all the above (when I am in a certain activity I am unavailable)?

Presence & Social Relatedness: Presence conveys a type of peripheral vision and social nearness (cognitively) that it is important to consider the "connectedness" implications of presence in terms of social networks and community.

Presence & attention: Both inward attention (managing access, interruptions) as well as outward attention (am I trying to get noticed, or to let someone know something, by setting my presence status to a certain state).

Presence & federation: Not just across various organizational boundaries but also across identity boundaries. Security (confidentiality, privacy) comes into play more prominently.

Presence of objects: presence of "things" as well as people so RFID is a type of presence. My presence can be inferred from objects near me or that I handle or wear (nTag).

Presence & agents/avatars: As technology progresses and we automate tasks via bots of various sorts, what type of presence do these "instances of me" have as they act as my proxy? Am I present in real-life when I am in Second Life?

I'll end with that one ... food for thought and for consideration as to how some of these items relate to assumptions currently made around presence systems. How many assumptions based on instant messaging, IP telephony and so on will get in the way of a more expansive view of presence? How do we deal with the discovering, aggregating, brokering, filtering, de-duping, syndicating and other issues? Are we wrong to assume only SIP-based presence systems need apply? Will we have a flat presence server model (federation, clearinghouses) or will we also need a server-of-server model? There's a lot to consider, from many angles - socially, organizationally (including business process and application aspects) as well as technically.

I've covered real-time collaboration since 1996 and for the past few years (at Meta and at Burton) have argued that presence needs to be considered as its own architectural topic.

antecipate: At your presence's convenience

A thoughtful and interesting assessment:

Recent studies on notification timing have highlighted four types of solutions to manage user interruptions: "immediate, negotiated, mediated, and scheduled". Interruptions can be delivered at the soonest (immediate), or the person has explicit control over when they will handle the interruption (negotiation). A third party system may also dynamically decide when best to interrupt the user (mediated), or always hold all interruptions and deliver them at a predefined time (scheduled). In most situations negotiation is said to be the best choice. But I believe adding feedback to presence will require exploring dynamic combinations of negotiation and mediation.

Source: antecipate: At your presence's convenience

The cost of distraction -- Alec Saunders .LOG

Alec has a compelling post on the proliferation of technology and its impact on productivity (the cost of interruptions). It includes a link to an article worth reading on how New Technology Takes Mental Toll on Workers. I commented on Alec's post, the comment is also posted below.

He talks about a productivity paradox, but let’s take that one step further. Today’s information driven workplace and the attendant technology, is a fundamentally disempowering environment. How is it reasonable to expect any work of consequence to be accomplished in that environment?

Speaking from personal experience, my highest productivity is on airplanes — no internet, no cell phones, no text messages.

Source: The cost of distraction -- Alec Saunders .LOG

Comment:

Alec, I agree with the notion that additional communication channels and pervasive connectivity give rise to significant interruption issues that impact productivity. A focus on attention management is relevant and deserves serious attention by IT strategists. So I am in total agreement on the need to better manage and triage our signal/noise challenges.

But on the other hand, what has been missing in the discussion, or so it seems, is a focus on the good side of this (it’s not all bad). If someone gets salient insight from me it that is necessary at the moment and it makes them more productive, then we are faced with a value decision. The problem is there is no good method in place to help parties make that value decision. Technology support is rudimentary so we often rely on social contracts between people (it’s ok for Jane Doe to bother me because she’s on my team but not John Doe because he’s in marketing and I know he takes forever to get to the point).

What is an interruption to me or to you could be valid, but the information or conversation that results might factor into the closing of a deal on the part of the requestor (or some other net-positive event). If electronic interruptions waste 28 billion man-hours per year in this country, at a cost of $588 billion as Basex points out, what would we think if we discovered that those interruptions resulted in 30 billion man-hours of productivity gain (on the part of those doing the interrupting) and garnered 600 billion in revenue (greater than the costs to those of us being pestered)?

I don’t have the answer. But we all seem to be going down this interruption-is-bad path when I believe the issue is more complex and involves a lot more than technology.